Can economic growth continue without fossil fuels? The IPCC thinks so — here’s why its decarbonisation models are broken

By Graham Palmer

(Source: VICE)
Published by INSURGE INTELLIGENCE, a crowdfunded investigative journalism project for people and planet. Support us to keep digging where others fear to tread.

In this third contribution to our symposium, ‘Pathways to the Post-Carbon Economy’, biophysical economist Graham Palmer assesses the plausibility of conventional economic forecasts of a global energy transition away from fossil fuels.

Most forecasts paint a rosy picture of continued, unimpeded economic growth, even as the world weans itself entirely off carbon-intensive energy sources. But are such scenarios really possible?

Palmer argues that they aren’t — not when we consider how the economy is fundamentally embedded in its biophysical environment. And if it isn’t, then we need a new approach to modelling, which pays greater attention the intimate relationship between energy, our societies, and their economies.


Climate change discourse is structured around competing narratives — degrowth, pro-renewables, pro-nuclear, localism, green business, techno-optimism, and so on. The energy scenario modelling by the UN’s Intergovernmental Panel on Climate Change (IPCC) provides a foundation for much of the discourse.

Integrated models are connected with socioeconomic and technological storylines to forecast a picture of key characteristics of future transformation pathways.

When we look at models from the most recent IPCC report (AR5), we see that that baseline scenarios project a 300% to 800% increase in GDP-per-capita by 2100. In these scenarios, strong mitigation is achieved with global consumption losses of only between 3 to 11% relative to baseline. Hence, the net-cost of decarbonising seems trivial over the long run.

From this perspective, the solution that follows is to put in place appropriate policies, support technology, remove fossil fuel subsidies and apply a modest but comprehensive carbon price.

The IPCC scenario modelling seems to gives weight to a green business, techno-optimist narrative of unimpeded green growth. Proponents of degrowth are marginalised and ‘peak oil’ is history. Yes, it will require large expenditures and continued research, but, according these models, we have good reason to be confident that low-carbon alternatives and efficiency can readily substitute for fossil fuels.

But how do we distinguish between model projections and model inputs — the accuracy of the values assumed to make the projections? Does economic growth emerge as an outcome of decarbonization models, or is it simply an assumed input? How do models connect growth to energy?

My research shows that the IPCC’s low-emission future scenarios are essentially superimposing energy transition models over standard macroeconomic growth theory, in a way that was consistent with 20th century economic growth.

What this means is that these key factors — economic growth, productivity improvements, and a reduction in energy intensity — are given, or rather, simply assumed as model inputs.

They do not emerge out of models, they are given to models.

The underlying assumption is that the macroeconomy of the next 100 years will be the same as the last 100.

Two percent of per-capita annual growth compounded over a century gives a 7-fold increase. Three percent gives 18-fold.

But if we consider that we are already at, or exceeding the ecological footprint, and that resource depletion is a one-way journey, how do we reconcile these growth projections with a sustainable future?

The 20th century was a remarkable period in human history.

Despite two major wars and global turmoil, the developed world advanced the standard-of-living of its citizens at an unprecedented pace.

The earlier development of coal fired steam power broke the hard organic limit of agrarian societies. The first Industrial Revolution saw the development of the steam engine, cotton gin, railroad, and modern manufacturing.

The second Industrial Revolution saw an epochal suite of inventions, including the internal combustion engine, electric motors and grids. Oil enabled mass transit and air travel, the Ford production system diffused through manufacturing.

This was an extraordinary period of human development underpinned by surplus energy flows. Is it being repeated?

Axiom: The nature of exponential growth is that the proceeding advances must match the preceding, not in absolute terms but in relative terms, meaning much higher absolute growth.

Conventional economists assume that growth emerges inevitably out of innovation and ideas — it doesn’t matter that it is difficult to imagine a much wealthier future, it is a law of nature that productivity will inexorably rise.

But it turns out that much economic growth has been enabled by surplus energy — possibly two-thirds of the growth that is usually attributed to ‘technical change’ is actually due to the availability of cheap energy, much of it easily transportable liquid fuels.

This means that the sort of growth figures that are assumed to be a ‘stylised fact’ of models will require radical improvements in energy supply technologies and energy end-use services.

When the assumption of continued growth in the face of climate change is relaxed, the cost of mitigation can increase by orders of magnitude.

Energy scenario modelling tends to adopt the high-end estimates of fossil fuel availability given in the Global Energy Assessment that assume that resource availability will not be a constraint on economic growth.

The biophysical perspective is that modern civilisation should be treated as a complex, intertwined system that is underpinned by dissipative energy and material flows. Economic growth is contingent upon energy that is available to drive material flows through the system.

The evolution towards service-based economies in the advanced economies has reduced the energy intensity of GDP but has not fundamentally altered this relationship. It is the net-energy, or the energy available after deducting the energy-industry-own-use and life-cycle embodied energy from the gross energy flows, that is the defining metric for understanding the link between energy and society.

How does the assumption of rising energy surpluses relate to energy sources that are assumed to increase over the next century?

Chris Martenson reports $70 billion worth of bankruptcy filings across 100 US shale oil companies, and Steve St Angelo reports a negative cash flow of $20 billion in 2017.

The countries with the highest penetration of wind and solar per-capita tend to have the highest residential electricity prices despite falling per-kW installation prices. The cost overruns of the four Westinghouse AP1000 nuclear reactors — two each in Georgia and South Carolina — has contributed to the collapse of Toshiba and thrown doubt on the completion of the projects. The Energy Return on Investment (EROI) of biofuels is substantially below that of oil.

Insight: It should be clear that energy is getting harder, not easier. In the absence of radical innovations, such as portable cold fusion devices, we cannot rely on cheap energy to drive the sort of one-off productivity improvements that we saw in the 20th century. We need to come to terms with the fact that economic growth may be incompatible with abatement targets.

In conclusion: we simply don’t have the understanding of the complexity and feedback processes to confidently project several decades into the future.

The IPCC’s grafting of 20th century growth and productivity data onto future scenarios completely misses the intimate relationship between energy and economies.

The downward trend of EROI is critical to understanding future standard of the living.


Graham Palmer is a PhD candidate at the Australian-German Climate and Energy College, University of Melbourne, and an engineer at Paltech Corporation. He is also on the scientific advisory committee of the International Society for BioPhysical Economics. He is the author of Energy in Australia: Peak Oil, Solar Power, and Asia’s Economic Growth (Springer, 2014).